tropical cyclone assessment of mexico - fairco · increase in tropical cyclone activity in the...
Post on 14-Aug-2020
1 Views
Preview:
TRANSCRIPT
1 Flood and Drought Assessment of Brazil | TransRe White Paper
IMPLICATIONS FOR LOCAL INSURERS
by James Rohman | December 2013
Tropical Cyclone Assessment of Mexico
1 Tropical Cyclone Assessment of Mexico | FAIRCO White Paper
. Figure 1. Mexico by state, for reference.
Figure 2. Landfalling hurricanes (all categories) by state, 1951-2000.
2 Tropical Cyclone Assessment of Mexico | FAIRCO White Paper
Executive Summary
Mexico’s Atlantic and Pacific coasts have both been affected by more
extreme and destructive tropical cyclones in the past 15 years.
This increased destructiveness has coincided with socioeconomic growth
along Mexico’s coastlines.
Pacific hurricanes make landfall twice as often (18%) as Atlantic hurricanes
(9%).
The combination of increased risks and increased insured values makes it
ever more important to understand future trends and developments in
tropical cyclone climatology.
The probability for tropical cyclones along Mexico’s coasts can be analyzed
by evaluating certain climate events.
The El Niño phase of the El Niño Southern Oscillation (ENSO) is commonly
associated with more tropical cyclones in the eastern Pacific basin, and less
tropical cyclones in the Atlantic basin.
La Niña conditions are associated with the opposite – fewer Pacific cyclones
and more Atlantic cyclones.
When the Madden-Julian Oscillation (MJO) produces westerly wind
anomalies, this increases the chances of a tropical cyclone in the eastern
Pacific and Gulf of Mexico basins by four times.
Easterly wind MJO anomalies tend to produce less favorable conditions for
tropical cyclone genesis.
The positive phase of the Arctic Oscillation (AO) increases the likelihood of
an active Atlantic hurricane season due to low surface pressure.
The negative phase of the AO is associated with a less active Atlantic
hurricane season.
These climate events can be tracked and monitored on a variety of
timescales to aid in the understanding of tropical cyclone climatology.
TransRe is working to support insurers to better understand the causes and
implications of the eastern Pacific and Atlantic hurricane seasons.
3 Tropical Cyclone Assessment of Mexico | FAIRCO White Paper
Introduction While extreme weather can cause widespread disruption and loss,
the science of climatology is still evolving, and the correlations are not
fully understood.
However, insurers are in the business of risk management and protecting
insurance buyers from their exposures. It is our job to understand and quantify
the linkages as best we can. While society (and we) may hope for the best, we
must also prepare for the worst. This paper will seek to highlight general
developments in the understanding of tropical cyclone climatology as it
affects both coasts of Mexico.
Mexico has Latin America’s second largest economy, with great potential
for further growth. The economy has displayed resilience in the face of a global
economic slowdown. The government has active plans to boost the country’s
competitiveness and growth, particularly in the energy production and
telecommunications sectors and 2014 projections are encouraging.
Mexico is susceptible to natural disasters along its Atlantic and Pacific
coastlines. Both coasts are also more heavily populated and urbanized than
they have been in the past, thanks to tourism and other economic activity.
Mexico is particularly vulnerable to rainfall from hurricanes and tropical
storms along the coasts - orographic lifting and adiabatic cooling from the
Sierra Madres intensify local rainfall. To understand the causes,
the first section of this paper summarizes the science behind both.
In recent years, Mexico has suffered a rise in the frequency and severity of
tropical cyclones. The second section of this paper summarizes the rationale.
Public perception of these risks is changing. Combined with increasing private
wealth, this will result in increased insurance penetration. Mexican insurers
must understand the impact of the changing risk profile to their exposures. The
final section of this paper highlights developments in this area.
4 Tropical Cyclone Assessment of Mexico | FAIRCO White Paper
The Science of Mexican Tropical Cyclones
The major factors driving Mexico’s tropical cyclone exposure are the El Niño
Southern Oscillation, Madden-Julian Oscillation and Arctic Oscillation.
El Niño Southern Oscillation (ENSO)
In cycles of 2-5 years, anomalous warm (El Niño phase) and cold (La Niña
phase) patches of ocean water form off western South America. They affect
60% of global weather patterns, including flood/drought patterns in Brazil,
USA, Africa, Indonesia, Australia and Southeast Asia.
Madden-Julian Oscillation (MJO)
A band of low-pressure develops in the Indian Ocean, travels east to dump
rain on SE Asia and Australia, and continues east as an anomalous weather
band. It is associated with large-scale variations in atmospheric/oceanic
conditions, affecting droughts, monsoons and tropical cyclones.
Arctic Oscillation (AO)
AO is an index of non-seasonal sea-level pressure north of 20oN. The effects
are explained by NASA climatologist Dr. James E. Hansen.
"The degree to which Arctic air penetrates into middle latitudes is related to the
AO index, which is defined by surface atmospheric pressure patterns. When
the AO index is positive, surface pressure is low in the polar region. This helps
the middle latitude jet stream to blow strongly and consistently from west to
east, thus keeping cold Arctic air locked in the polar region. When the AO
index is negative, there tends to be high pressure in the polar region, weaker
zonal winds, and greater movement of frigid polar air into middle latitudes."
AO influences the entire northern hemisphere, affecting North American and
European populations, and also has a strong influence on tropical activity in
the Atlantic basin.
Interaction between MJO and ENSO
MJO is highly variable from year to year, with periods of strong activity followed
by weak or non-existent conditions. The atmospheric science community
suspects that MJO is partially linked to the ENSO cycle. Typically, strong
MJO activity is observed 6-12 months before the onset of El Niño, but is
virtually absent during the maxima of some El Niño episodes. Meanwhile, MJO
tends to be more active during La Niña periods. MJO can influence the
ENSO cycle by contributing to the speed of development, and perhaps
the overall intensity of El Niño episodes. In this way, MJO modulates
ENSO. It is important to track both as they will affect global weather patterns.
5 Tropical Cyclone Assessment of Mexico | FAIRCO White Paper
Average number of
El Nino
(20 events)
Neutral
(19 events)
La Nina
(13 Events)
TCs per ASO 6.9 8.42 9.3
Hurricanes per ASO 4.25 5.47 5.8
Landfalling TCs per ASO 1.34 2.32 2.87
Landfalling hurricanes per ASO 0.6 1.16 1.92
Increased Rainfall & The Sierra Madres
Much of the damage from hurricanes and tropical storms comes from the
associated rainfall. As evidenced by 2013 Hurricane Raymond, 12-16 inches
of rainfall can accumulate on the ocean-facing slopes of the Sierra Madres.
This is due to orographic lift and adiabatic cooling. Orographic lift occurs
when air masses are forced up to high elevations as they move over
terrain. When the air mass gains altitude, adiabatic cooling occurs: as
atmospheric pressure decreases, the internal air expands, the volume
increases and temperature falls. Dew point is attained and rainfall will drop
at a predictable altitude. These processes combine to intensify rainfall from
incoming tropical cyclones along the Mexican coasts.
Hurricane Climatology of ENSO
El Niño events are associated with an increase in eastern North Pacific
tropical cyclone activity, and a decrease in tropical cyclone activity in the
Atlantic basin. Conversely, La Niña events area associated with a
decrease in tropical cyclone activity in the eastern Pacific basin and an
increase in tropical cyclone activity in the Atlantic basin.
ENSO affects both the intensity and propensity for landfall in the Atlantic
Basin. During a La Niña phase, there is a 10% greater chance of tropical
cyclone genesis, and 24% greater chance of tropical cyclone landfall. During
an El Niño phase, there is 18% less chance of tropical cyclone genesis, and
42% less chance of tropical cyclone landfall.
Table 1. The relationship of ENSO and Atlantic tropical cyclone activity, 1950-2000. (ASO= August-September-October)
El Niño increases Pacific and decreases Atlantic frequency because of its im-
pact on vertical wind shear; increased vertical wind shear acts to prevent
tropical disturbances from developing into more convective cyclones, as
displayed in Figure 3.
6 Tropical Cyclone Assessment of Mexico | FAIRCO White Paper
Figure 3. Effect of wind shear on a storm’s latent heat.
Tropical cyclone damage increases with the square of wind speed, so
more intense hurricanes translate to significantly more damage. In the US
(where more data is available) the average damage per storm is $1.6B in La
Niña years, but just $0.8B in El Niño years.
In the fifty years from 1951-2000, six major hurricanes struck Mexico’s
Caribbean coast. They developed from the warm tropical Atlantic waters east
of the Lesser Antilles during the cold/neutral La Niña phase. As they
entered the Caribbean, their strength intensified due to favorable La Niña
conditions. Most made double-strike landfall, hitting first on the Yucatan
peninsula and making a second hit on the Tamaulipas/Veracruz coasts. For
example, in September 1988, Hurricane Gilbert made landfall on the
northeast Yucatan Peninsula as a Category 5 with sustained winds of 150mph,
continued west-northwest across the Gulf and hit the south of Tamaulipas
State. Torrential rains accompanied the storm over the coast and into the
mountainous areas. Overall damage (including 60,000 destroyed homes)
was estimated at US$2B. Adjusted for inflation, that would US$3.8B in 2013
dollars. However, if Hurricane Gilbert hit the same regions today, storm
damages would be in excess of US$6B due to inflation and population
shifts/economic development.
It is important to note that ENSO isn’t the only climate factor to affect hurricane
development. Other environmental conditions to consider include pre-existing
vertical wind shear, sea surface temperatures, the Saharan Air Layer, and also
the Arctic Oscillation and Madden-Julian Oscillation.
7 Tropical Cyclone Assessment of Mexico | FAIRCO White Paper
Average number of
AO Negative
(bottom tercile)
AO Neutral
(middle tercile)
AO Positive
(top tercile)
TCs per ASO 7.18 7.53 9.24
Hurricanes per ASO 4.82 4.94 5.47
Landfalling TCs per ASO 1.82 1.76 2.94
Hurricane Climatology of Arctic Oscillation
AO is strongly related to tropical cyclone activity in the Atlantic basin (to date
its relationship with the Pacific has not been studied). 7 of the 10 most active
Atlantic hurricane seasons occurred during positive AO phases. Those
seven years averaged 13 tropical storms per season, compared to the long-
term average of 8.1. Conversely, 7 of the 10 least active years occurred
during negative AO phases, averaging just 4 tropical cyclones per year.
Approximately 29% more tropical storms occur during AO positive phases,
15% more hurricanes occur during AO positive phases, and the storms have a
62% greater likelihood of landfall.
Table 2. Relationship of Arctic Oscillation tropical cyclone activity (1951-2000).
(ASO=August-September-October).
Hurricane Climatology of Madden-Julian Oscillation
MJO strongly modulates tropical cyclone activity over the Gulf of Mexico and
western Caribbean Sea as well as the eastern Pacific Ocean. MJO causes
alternating periods of westerly and easterly wind anomalies over the tropical
ocean basins. Tropical cyclone activity tends to vary strongly in association
with these intraseasonal wind variations.
In the Gulf of Mexico, tropical cyclone formation during westerly MJO
phases is over three times more likely than during easterly MJO periods.
Hurricane genesis is four times more likely during westerly MJO periods.
The distribution of tropical cyclone genesis events indicates a strong
preference for Gulf of Mexico, western Caribbean and western Atlantic tropical
cyclones to form during westerly MJO events. Hurricanes of Category 3 or
greater also have a higher likelihood of occurring during the westerly
MJO phase.
Figure 4. Tropical cyclones formations by MJO phase (tropical storm tracks in blue, hurricane tracks in red, 1949-1997). MJO Westerly anomaly is on the left, and MJO Easterly anomaly is on the right.
8 Tropical Cyclone Assessment of Mexico | FAIRCO White Paper
Climate Event Phase Expected ForecastAssociated Climate
Conditions
El Niño
24% less storms in
the Atlantic and 5%
more storms in the
eastern Pacific.
La Niña
Increase in tropical
cyclones in the
Atlantic basin and a
decrease in storms
in the eastern Pacific.
Easterly wind
anomalies
Much less chance of
tropical genesis in
eastern Pacific and
Atlantic basins.
Less convective
activity due to
easterly winds.
Westerly wind
anomalies
Hurricane genesis is
four times more likely
in eastern Pacific.
Hurricane and
tropical storm
genesis are 4 and
3.3 times more likely,
respectively, in the
Gulf of Mexico.
MJO changes
atmospheric and
oceanic parameters
to favor tropical
cyclone genesis.
NegativeLess active Atlantic
hurricane season.
Higher surface
pressure in the
poles and weaker
zonal winds.
Positive Very active Atlantic
hurricane season.
Low surface
pressure in polar
regions.
Change in vertical
wind shear.El Niño Southern Oscillation
Madden-Julian Oscillation
Arctic Oscillation
During westerly MJO phases, hurricane genesis in the eastern Pacific is
over four times more likely to occur than during easterly periods. The
genesis tends to occur closer to the coast, increasing the likelihood of landfall.
Patterns can be identified from the historical record:
During the active 1995 Atlantic hurricane season (19 named
storms) 5 tropical storms formed in a 7 day period (August 22-28)
that coincided with a westerly MJO phase. During the following
month (August 29-September 27), only one tropical cyclone formed
during an easterly MJO phase.
A similar pattern was noted during the active 2008 season.
Summary of Conditions
Table 3. Summary of the conditions expected from each climate event and phase.
9 Tropical Cyclone Assessment of Mexico | FAIRCO White Paper
The Impact of Tropical Cyclones to Date The states along northwest (Sinaloa, Baja California) and south (Michoacan)
Mexico are most frequently hit by storms, as are the Yucatan Peninsula and
Tamaulipas on the Atlantic coast. In a study of 50 years of hurricane data (see
table below) it was noted that Pacific hurricanes are only 25% more frequent
than Atlantic hurricanes, but the rate of Pacific landfall is double that for
Atlantic hurricanes.
Table 4. Percentage of landfalling hurricanes, 1951-2000.
Since 2000, the number of landfalling tropical cyclones has increased,
exceeding the long term average in twelve of the past fourteen years. At the
same time, the number of people affected, and the extent of the economic
damage has also increased.
Table 5. Recent extreme tropical cyclone events in Mexico.
Even in quiet hurricane seasons, tropical development can be sparked at
almost any time. After a quiet start to the 2013 hurricane season, Tropical
Storm Manuel and Hurricane Ingrid struck southern and central Mexico
within 24 hours of each other, displacing thousands and creating widespread
damage. This was the first time since 1958 that two storms hit both coasts
within 24 hours, demonstrating the ever-present threat of extreme weather.
Region TotalHurricanes/
Year
Number of
Landfalls
Landfalls/
Yr%
Atlantic 294 6 27 0.5 9
NE Pacific 368 7 65 1.2 18
Landfalling
Cyclones
Estimated Economic
Cost ($B)
Estimated
Population
Impacted ('000s)
2000 2 0.1 30
2001 5 0.4 6
2002 3 0.84 509
2003 2 0.1 1
2004 1 NA 0
2005 4 8 2,970
2006 3 0.27 271
2007 4 0.6 203
2008 1 0.08 500
2009 2 0.04 112
2010 4 5.9 555
2011 5 0.1 550
2012 2 0.85 87
2013 4 0.95 70
10 Tropical Cyclone Assessment of Mexico | FAIRCO White Paper
Implications for Insurers (Models & Exposures) Climate Indices
It is apparent that ENSO, MJO and AO have distinct effects on hurricane
climatology on both coasts of Mexico. Each event corresponds to a greater
or lesser percentage of tropical cyclone genesis in the Pacific and Atlantic
Ocean basins. At TransRe, we track events and monitor forecasts from
government and private sources to keep the most up-to-date predictions
for each climate index.
Improving regional climate models is extremely important given the prospect of
increased frequency and severity of tropical cyclones. These costs have
already risen, with the 2013 losses being the most recent example, and this
trend can be expected to accelerate based on observable data.
Mexico’s insurance industry must be fully prepared for the financial impact of
these weather patterns. With an insurance penetration level of
approximately 1.1% of GDP, and the likelihood that insured losses from
tropical cyclones will grow significantly in the future, private insurers in
Mexico should continue to manage their exposures very carefully if they
are to actively contribute to the regeneration of the economy following
such events.
How FAIRCO Can Make A Difference
We can help in a number of ways. We work with customers and brokers to
design the right insurance program to protect balance sheets and income
statements. We are working to understand the changes that are driving the
increase in extreme weather events, and to incorporate predictive measures of
ENSO, MJO and AO into reinsurance calculations. The same research can be
used by insurers for catastrophe mitigation training and development of more
efficient products.
11 Tropical Cyclone Assessment of Mexico | FAIRCO White Paper
Glossary
Adiabatic Cooling- The cooling and subsequent precipitation of an air-mass
as it rises over a mountain range.
Arctic Oscillation (AO) - An index of the dominant pattern of sea-level
pressure variations in the Northern hemisphere.
Easterly winds - Winds that blow from east to west.
El Niño – Anomalously warm phase of ENSO cycle.
El Niño Southern Oscillation (ENSO) – The 2-5 year Pacific Ocean cycle
between El Niño and La Niña conditions that affects 60% of global weather.
La Niña - Anomalously cold phase of ENSO.
Madden-Julian Oscillation (MJO) – Major part of intraseasonal (30-90 days)
variability in the tropical atmosphere. Travels east from the Indian Ocean.
Orographic Lifting – The gain in altitude of an air-mass as it travels along a
mountain range.
Saharan Air Layer - An intensely dry, warm and occasionally dust-laden layer
of the atmosphere which can influence the cooler, humid surface air of the
tropical Atlantic Ocean.
Sierra Madre - The predominant mountain system in Mexico that runs the
length of the country.
Surface Pressure- Atmospheric pressure at a location on the Earth’s surface.
Westerly winds- Winds that blow from west to east.
Wind Shear – The rate at which wind velocity changes from point to point in a
given direction. High vertical wind shear results in decreased probability of
strong tropical cyclones.
12 Tropical Cyclone Assessment of Mexico | FAIRCO White Paper
Citations
Ezilon Maps.
Maloney and Hartmann (2002). ‘Modulation of Hurricane Activity in the Gulf of
Mexico by the Madden-Julian Oscillation’.
http://www.atmos.washington.edu/~dennis/Maloney%26HartmannScience.pdf
University of Illinois. ‘Interaction with El Nino’.
http://ww2010.atmos.uiuc.edu/(Gh)/guides/mtr/hurr/enso.rxml
Larson et al. (2005). ‘Characteristics of Landfalling Tropical Cyclones in the
United States and Mexico: Climatology and Interannual Variability’.
http://journals.ametsoc.org/doi/pdf/10.1175/JCLI3317.1
E. Jauregui (2003). ‘Climatology of landfalling hurricanes and tropical storms in
Mexico’. http://www.ejournal.unam.mx/atm/Vol16-4/ATM16401.pdf
Gottschalk et al. from NOAA. ‘Madden Julian Oscillation (MJO)’
http://www.cpc.ncep.noaa.gov/products/precip/CWlink/MJO/MJO_summary.pdf
13 Tropical Cyclone Assessment of Mexico | FAIRCO White Paper
Contacts
To discuss any of the contents of this paper, please contact:
James Rohman
Global Catastrophe Management
T: 1 (212)-365-2438 E: jrohman@transre.com
To discuss the implications of this paper, and your reinsurance needs in
response, please contact:
Javier Vijil Juan-Carlos Roa
President, Latin America Senior Vice President and Underwriting Manager, Latin America
T: 1 (786)-437-3901 T: 1 (786)-437-3944
E: jvijil@transre.com E: jroa@transre.com
Marco Giovane
Vice President and
Marketing Manager, Latin America
T: 1 (786)-437-2902
E: mgiovane@transre.com
14 Tropical Cyclone Assessment of Mexico | FAIRCO White Paper
Copyright and Disclaimers
The material and conclusions contained in this document are for information
purposes only and the authors offer no guarantee for the completeness of its
contents. The statements in this document may provide current expectations of
future events based on certain assumptions. These statements involve known
and unknown risks, uncertainties and other factors which are not exhaustive.
The authors of this document undertake no obligation to publicly revise or
update any statements, whether as a result of new information, future events
or otherwise and in no event shall Fair American Insurance and Reinsurance
Company or any of its affiliates or employees be liable for any damage and
financial loss arising in connection with the use of the information relating to
this document.
About FAIRCO
FAIRCO is the logo utilized for Fair American Insurance and Reinsurance
Company and its subsidiary(ies). Fair American Insurance and Reinsurance
Company is a wholly-owned subsidiary of Transatlantic Reinsurance
Company.
TransRe is the brand name for Transatlantic Holdings, Inc. and its subsidiaries
(including Transatlantic Reinsurance Company).
Visit www.fairco.com for additional information.
Copyright © 2015 Fair American Insurance and Reinsurance Company. All
rights reserved.
Fair American Insurance and Reinsurance Company
One Liberty Plaza, 165 Broadway
New York, New York 10006
top related